CONTINUOUS RESISTIVITY PROFILING IN SHALLOW

MARINE AND FRESH WATER ENVIRONMENTS

D.D. Snyder, Scott C. MacInnes, M.J. Raymond, and Kenneth L. Zonge

Zonge Engineering and Research Organization
Tucson, AZ
SkipS@zonge.com

ABSTRACT
In this paper, we describe an instrument system for performing continuous resistivity profiling in
shallow freshwater and marine environments. Using a streamer cable containing 9 electrodes, the
system continuously samples the dipole-dipole resistivity at n-spacings 1 through 6. The system can be
installed aboard a variety of small inboard or outboard powered vessels in a few hours. Hand-held or
marine GPS units provide location information that is recorded by a laptop computer. With this system,
up to 40 line-km/day of dipole-dipole data have been collected. The resistivity data are merged with the
GPS positions as a post-processing step. The final step in the post-processing is the inversion of
overlapping segments of each profile using a 2-D smooth model. The inversions provide high resolution
images of the geoelectric cross-section. The depth of investigation ranges from 20-30 m, with a 10 m
dipole spacing. Over the last 4-years, we have performed surveys on the Ohio River, near Louisville,
KY, on tidal estuaries and bays along the Atlantic coast in Delaware, Maryland, Virginia, and North
Carolina, and in Tampa Bay, Florida. Data from these surveys will be used to illustrate the final
deliverable from a survey.

INTRODUCTION
The electrical resistivity or conductivity of the shallow sub-surface is of particular interest to
environmental geophysicists. It is a physical property directly affected not only by the presence of
conducting foreign material that may have been improperly disposed of, but also by the chemistry of the
saturating fluids. Environmental hydrologists are sometimes able to use resistivity measurements to
map contaminant plumes as they migrate down the hydraulic gradient from their source. Moreover,
hydrologists know that fluid migration within our aquifers is an important source of contamination in
our lakes, rivers, and tidal estuaries. Thus, resistivity measurements can be valuable for investigation of
shallow ground-water hydrology from the waterside as well as the landside.
Four years ago, Zonge Engineering assembled an instrument system to conduct a towed-array
resistivity survey on the Ohio River near Louisville, KY. Since that time, we have had the opportunity
to conduct a number of other surveys with that system, in both shallow freshwater and shallow marine
environments. In the main, these surveys have been conducted in support of groundwater and
environmental investigations. However, our experience suggests that these surveys may be effective in
mapping or locating conductive features such as buried culture and, possibly, UXO. Our purpose here is
to describe our system for measuring resistivity and IP in shallow water, and the data processing
involved. We will also share in a general way how these data are currently being used. Several other
papers being presented at this years SAGEEP speak more directly to specific applications.

1
 BASIS FOR RESISTIVITY MEASUREMENTS IN SHALLOW WATER
Qualitatively, resistivity measurements in shallow water can be viewed in much the same way as
borehole resistivity measurements. In the borehole environment, the borehole fluid causes a departure
of observed resistivity from the true formation resistivity. In an otherwise uniform space, making the
electrode array spacing large when compared with the borehole diameter can make the resistivity
departure negligibly small. Well log analysts have calculated departure correction charts for many types
of electrode arrays (e.g., normal array, lateral log, etc). The departure curves facilitate correcting
resistivity readings for the effects of the borehole annulus and for effects of thin beds (i.e., beds whose
thickness is not large compared with the electrode spacing).1 Similarly, we expect that the departure
of the measured resistivity from the resistivity of the sub-bottom will be relatively small, provided
electrode array dimensions are relatively large with respect to water depth.
We have chosen to use the dipole-dipole
array for our measurements because it is a 4-
electrode array that requires no remote (i.e.,
infinite) reference electrodes and is therefore easy
to implement on a towed cable. Moreover,
provided a multiple-channel receiver is available,
receiver dipoles at different n-spacings can be
measured simultaneously and so affect a
combination of profiling and sounding. The latter
reasoning, it seems to us, provides a compelling
reason for using the dipole-dipole array for all of
our marine surveys.
We have confirmed and to some extent
quantified this assumption by computing a set of
departure curves for the in-line dipole-dipole array
measured at the surface of a 2-layer earth for
integer n-spacings ranging from 1-6. Figure 1
shows sets of these curves computed for n=1, and
n=6, respectively. The dipole-dipole array
geometry is illustrated in the Figure insets. In the
Figure, the abscissa is the thickness of the top layer
expressed in units of the dipole spacing (a). The
ordinate is the observed or apparent resistivity
normalized by the true resistivity of the top layer Figure 1: Resistivity departure for the inline dipole-dipole
(r1). The 4 curves represent different sub-bottom array over a 2-layer geometry.
2
resistivities. These curves show that the departure
of the observed apparent resistivity from the true
resistivity of the sub-bottom is no larger than 50 percent when the water depth is less than a/2 for bottom
sediments saturated with the water representing the top layer (r2/r1=3-5).

1
Departure charts for normal and lateral array resistivity logs can be found in many of the older manuals published by major
well log service companies (e.g., Schlumberger [1]1. Schlumberger, Log Interpretation Charts. 1972, Schlumberger
Limited.).
2
The formation factor (F) that is typically associated with unconsolidated sub-bottom (sand/silt) is 3 to 4.

2
 On the basis of departure curves such as those in Figure 1, and catalogs of three-layer curves
published by Charles Elliot [2], we fabricated our first streamer cable in the summer of 1997 with
electrodes spaced at 10-m intervals. With this cable, therefore, we can feel assured of being able to
adequately measure the resistivity of the sub-bottom in water depths of up to 5 m or more. Figure 1a

Figure 2: Functional block diagram of marine resistivity/IP Figure 3: Photograph showing installed system
system aboard a 22-ft Boston Whaler.
shows that with a water depth of 5 m it would be difficult to resolve a change in the sub-bottom
resistivity from 10 to 30. However, changes at such high resistivity contrasts are easily resolved at the
larger n-spacings.

INSTRUMENT SYSTEM
Figure 2 shows a block diagram of the marine system that we have deployed to conduct the
shallow water resistivity surveys. The photograph in Figure 3 shows one of the authors with the marine
resistivity/IP system installed for one of our recent surveys. The small photograph inset into Figure 3
shows the survey vessel for this installation. As Figure 2 indicates, the system is comprised of 4 main
sub-systems:
1. Streamer Cable
2. Navigation
3. Resistivity/IP Measurement
4. Data Acquisition

Streamer Cable Sub-System

The streamer cable consists of a multi-conductor cable that is approximately 120 m long. Nine
take-outs have been constructed at 10 m intervals using a 3M (PN 82-F2) vulcanizing splicing kit. At
each of the take-outs, one of the cable conductors is brought through the take-out splice and terminated
with a weather-proof socket connector (Grote PN 66160).3 The electrode is formed using a length of
tinned copper tubular braid that has been slipped over the cable and positioned at one of the takeouts.
We used a length of foam pipe insulation to cover the takeout splice and cable and to provide a backing
material so that the copper braid expands to its maximum diameter that is considerably larger than the

3
Weatherproof automotive connectors can be used for streamers meant to float on the surface. More expensive waterproof
connectors are more appropriate for submerged cables.

3
streamer cable or the 3M splice. Photos of the electrodes
are shown in Figure 4. For this cable, we enlarged the
current electrodes (Figure 4, bottom) to reduce their
contact resistance for surveying in freshwater. This is not
necessary if the survey is to be conducted in salt water
since the contact resistance for the smaller potential
electrodes (electrode numbers 3-9 in Figure 2) is
sufficiently low.

Navigation Subsystem
The navigation subsystem consists mainly of a
differential quality GPS system (< 1m position accuracy).
As suggested in the system block diagram in Figure 2, it
is highly desirable that the navigation system be
interfaced with a digital Fathometer. However, we have
not always had such a Fathometer available. We have
used many different types of GPS receivers including
marine differential units (Coast Guard or U.S. Army
Corps of Engineers radio correction broadcasts), satellite
corrected GPS receivers such as the Trimble (Trimble
LandStar), and a militarized version of a hand-held GPS
receiver. Some of these receivers have had their own Figure 4: Potential (top) and current electrodes.
data loggers and have been used to store navigation fixes
and the corresponding GPS time at 1-sec intervals. At
other times, we have been able to configure the receiver
to transmit periodic NMEA sentences containing time, position, and sometimes speed and heading over
a serial RS-232C port. These sentences are captured on a spare COM port on the Data Acquisition
computer and stored to its hard disk.
Some of the better marine GPS systems are integrated with a Fathometer and will record water
depth directly. However, we have installed our system aboard only one vessel where this was the case.
The system shown in Figure 1 included a separate Fathometer with serial output. A commercially
available hydrographic survey package (HypackTM - Coastal Oceanographics, Inc) was used to merge
the fathometer measurements with the appropriate navigation fixes and then both retransmit it to our
data acquisition computer and store the data on the laptop computer that was running Hypack. In that
installation, the customer supplied both the vessel and the navigation sub-system.
Resistivity/IP Measuring Subsystem
The resistivity/IP measuring sub-system consists of a Zonge GDP-32/32II receiver with 7 analog
channels. The transmitter has an output power rating of approximately 600W and is powered with a
primary supply of 24Vdc. In fresh water where contact resistance is significantly higher than in salt
water, we employ a DC-DC Power Booster that can step up the 24Vdc primary voltage to voltages as
high as 400Vdc. The receiver is connected to the 6 receiver dipoles (i.e., M1-M2, , M5-M6 in Figure
2). A 7th analog channel is used to monitor a voltage proportional to the output current (typically 1V/A).

4
 Operating with special firmware, the receiver performs a periodic resistivity/IP measurement and
stamps it with the time of the reading using the receivers precision real time clock.4 These data are
stored both internally in a battery-backed data cache and transmitted out the receivers COM port for
capture by the data acquisition computer. The times recorded with the resistivity and navigation data
form the basis for associating a position with each of our measurements.
Data Acquisition Subsystem
Although the navigation and the resistivity/IP measuring subsystems generally record their data
independently, a few steps in the post-processing of data are greatly facilitated if not totally eliminated
when the navigation data are captured on a single acquisition computer. We are using a laptop computer
with 2 COM ports for acquisition. At this point, our acquisition software can easily be described as ad
hoc. But with each survey, we develop new or modified software capabilities both for acquisition and
for post-processing and we look forward in the future to having a fully integrated system for acquiring
these data and performing some of the necessary processing in real time. Among other things, we hope
that new versions of our acquisition software will implement a real-time water-fall display of the 6
resistivity and 6 IP traces we acquire.

DATA PROCESSING
The system described in the previous section can easily acquire 30-40 line-km of dipole-dipole
data per day when surveying long lines. Data are usually acquired at 4-sec intervals using surveying
speeds of 4-5 km/hr. This means we acquire a set of 6 resistivity and IP measurements at approximately
5-m (half-dipole) intervals. This is a huge amount of data to process. Our system for processing these

Figure 5: Flow diagram showing major steps in processing marine resistivity/IP data.

4
The standard Zonge GDP-32/32II receivers have timing circuitry and a real time clock driven by an ultra-stable quartz
oscillator. The oscillator is also used to maintain local synchronous phase references for all binary frequencies from 2-10
Hz to 214 Hz. Once the oven-stabilized crystal has been warmed up, time drift is less than a millisecond per day. The real
time clock keeps time to the second. A recent upgrade to the clock hardware permits the clock to be read to the nearest
1/32 sec.

5
data has evolved over a period of more than 4-years. The
Location Map
flow diagram in Figure 5 shows the major steps in the Line D
processing flow. The major steps include: 27.85

1. Basic Processing
2. Model Inversion
3. Model Display

Latitude (degrees)
Basic Processing

9
18

0
41
Two main functions are accomplished in the data

9
8
2

41
65
11

90
27.84
reduction step. Firstly, we merge the navigation data with the
resistivity data and thereby obtain data files that relate vessel
position along the survey track (i.e., station number) to a
geographic position. Secondly, we estimate the position of
each of the 9 electrodes in our streamer cable and use these
positions to calculate apparent resistivities.
Navigation Reduction - Because of the variety of navigation 27.83
systems with which we have had to contend, we generally run -82.41 -82.40
Longitude (degrees)
our navigation file through a reformatting program (box 1 in
Figure 5) to reformat the navigation data and present it in a Figure 6: Portion of track plot.
standard format to the main reduction program, GPSIP (box 2
in Figure 5). Using the navigation data, GPSIP computes an accumulated station number starting at
the beginning of the line. The station number is the accumulated distance along the profile expressed in
appropriate distance units (usually meters or feet). Among other things, GPSIP produces a file that
tabulates station number and
geographic position. We use these
data either graphically or
numerically to relate a profile plot
point expressed as a station number
to a geographic position. Figure 6
is a graphic representation of a track
station plot.
Resistivity Reduction and Plot
Point Estimation Resistivity
reduction has evolved extensively
over the 4+ years that we have been
running these surveys. In our
original survey, we assumed that
the streamer cable was in line with
the survey vessels heading as
illustrated in Figure 7a. Using the
coordinates of the GPS antenna, the Figure 7: Figure showing geometrical relationship for estimating streamer
geometry of the cable and its tie-off electrode position.
point on the vessel, we used basic
analytic geometry to compute the geographic positions of each of the 6 plot points for the dipole
measurements. This works well for surveys with long straight tracks, but proves unsatisfactory when
the survey vessel changes course more than a few degrees while surveying. We have called our early

6
assumption the Rigid Tail assumption, and Figure 8 illustrates the consequences of that assumption
when it is applied to a profile where there have been significant course changes. The Figure shows the
ships GPS positions as small circles. It plots the estimated position of the 6 dipole-dipole plot points as
a sequence of 6 equal-spaced + symbols. This plot shows how the Rigid Tail reduction creates a fan

6600
Northing (feet)

6500

circle: GPS location

plus: electrode
(Vertical scale exaggerated)

6400

3300 3400 3500 3600 3700 3800 3900 4000 4100 4200 4300 4400 4500 4600 4700 4800 4900 5000

Easting (feet)
Figure 8: Track of survey ship with electrode locations estimated with Rigid Tail assumption.

effect during a course change. Where it is severe, the fan effect shown in Figure 8 can significantly
distort our efforts to model and display a resistivity cross-section.
We have subsequently modified our
Rigid Tail assumption to a Curved Tail Station(m)
1200

1000

assumption. We assume that under gentle 800

600

400

200

0
maneuvering, the streamer cable will 0 10
follow the ships track as illustrated in
2
Figure 7b. The station number of each
Depth
electrode and plot point is used to 4

Resistivity(ohm-m)
Rho1
interpolate the geographic coordinates from
Depth(m)

6 Rho2
the GPS track. The observed voltage Rho3
8
differences are reduced to apparent Rho4
Rho5
resistivity based on four sets of electrode 10 1
Rho6
coordinates that are not necessarily co- 12
linear. Figure 9 is an example of data that
14
have been reduced with GPSIP.
Figure 9: Portion of resistivity profile plot.

Model Inversion
Our data processing steps normally include the inversion of the apparent resistivity profiles using
a 2.5D smooth-model inversion program described by MacInnes and Zonge [3]. This program has been
modified to invert dipole-dipole pseudo-sections with as many as 150 dipoles. Since our profiles are
typically many kilometers long, we divide the profile into sections of approximately 1200m that overlap
with adjacent sections by at least 100m. Figure 10 shows the model section, the calculated resistivity,
and the observed resistivity for one section of a profile acquired during a recent survey in a shallow
marine environment in Florida.
Although the data for Figure 10 included Fathometer information, we are not yet able to include
the Fathometer profile into our resistivity model and thus constrain the model. However, we are
currently working to implement a modification to our inversion program that will allow us to

7
incorporate that information. We believe that this modification will considerably improve the image
generated by the inversion process.

Figure 10: 2.5D Smooth-model Inversion results.

Model Display
The inversion results are output into data files that are compatible with either SURFER (Golden
Software) or Oasis Montaj (GeoSoft). The model cross-section in Figure 10 was generated using
SURFER 7.0.

APPLICATIONS
Most of the interest in applying these towed-array resistivity measurements has come from the
ground-water community. Several papers presented at this symposium present data acquired using the
Zonge system. In connection with hydrology, estimates of the resistivity of the sub-bottom can be used
as a basis for estimating the resistivity (hence salinity) of the pore water. In shallow estuaries,
freshwater often seeps into the estuary from aquifers that crop out down gradient from nearby land
areas. Changes in the pore water resistivity cause a significant increase in the resistivity according to
well-known empirical laws governing the resistivity in porous rocks. Archies law [4] relates bulk
resistivity of a porous rock to a Formation Factor according the formula

T = f F ; F = m (Archie's Law) (1)

Where
rT= The bulk resistivity of the saturated material
rf = The resistivity of the saturating fluid

8
 F = The formation factor
f = Fractional porosity of matrix material
m = cementation exponent (for unconsolidated sands ~2)
Archies law suggests that the bulk resistivity of an unconsolidated sand/silt layer in a shallow
estuary might be up to 4 times the resistivity of the seawater (typically about 0.3-0.4 ohm-m). Thus, in
the absence of fresh or brackish water in the subsurface, we expect a resistivity in the sub-bottom of
about 1-1.5 ohm-m. Notice in the model cross-section shown in Figure 10 that the model resistivity in
the zone 10-15m deep is more than 20 ohmmeters. Unless there is a significant change in the formation
factor of the sub-bottom, the model resistivities suggest that the water saturated the sub-bottom has a
resistivity of 5 ohmmeters or more (border-line fresh water).
Lest we forget, this system is also capable of measuring the IP response. Although our
customers have been primarily interested in ground-water applications, there are potential applications
for using the IP measurements. Wynn [5] has observed a measurable IP response associated with
titaniferous beach sands. The response is small and requires that the streamer be submerged so that it is
closer to the sub-bottom. That material containing small amounts of disseminated metallic
mineralization produces an IP effect comes as no surprise to mining geophysicists. But IP effects are
also associated with massive sulfides and with smaller but still finite-sized metallic objects. Indeed, one
of the earliest applications of the IP method was the Mark 5 Beach Mine Detector developed and
deployed briefly during World War II [6]. Additional anecdotal comments have appeared in the
literature suggesting that IP measurements may be useful for detecting metallic objects resting on the
water bottom or in the near sub-bottom.

CONCLUSIONS
We have described a system for acquiring resistivity and IP data from an electrode streamer
towed behind a small outboard motor-powered vessel. Assembled from commercially available
instruments for measuring resistivity and IP, the system can be installed aboard a small survey vessel in
a few hours and is designed for deployment in shallow water of depth 5-10m. . It can acquire dipole-
dipole data (n=1-6) at the rate of 10s of line-km per day. We have deployed this system primarily in
connection with ground-water investigations. However, we expect that other applications will emerge
as environmental scientists become acquainted with the economics of the deployment of towed-array
resistivity measurements. A key advantage to the adaptation of the common geophysical techniques for
deployment in shallow water is the access that these water bodies provide into areas that are otherwise
inaccessible from the landside. As geophysicists, we often find that geophysical surveys along
shorelines are rendered difficult or impossible because of complex land ownership and the cultural
effects (e.g., fences and buried utility lines). But in a small craft, we are free to float right by the end of
the dock.

9
 REFERENCES

1. Schlumberger, Log Interpretation Charts. 1972, Schlumberger Limited.

2. Elliot, C.L., Theoretical Response-Three Layered Earth. undated, Elliot Geophysical Company:
Tucson, AZ.
3. MacInnes, S., and Ken Zonge. Two-dimensional Inversion of Resistivity and IP Data with
Topography. in 102nd Annual Northwest Mining Association Convention. 1996. Spokane, WA.
4. Archie, G.E., The electrical resistivity log as an aid in determining some reservoir
characteristics. AIME Tech. Paper 1422, 1942.
5. Wynn, J.C., Titanium geophysics: The application of induced polarization to sea-floor mineral
exploration. Geophysics, 1988. 53(3): p. 386-401.
6. Grow, L.M., Induced Polarization for Geophysical Exploration, in The Leading Edge. 1982. p.
55-70.

10